3D object clouds, first introduced by Hong and Brooks, visualize the pairwise similarity between a set of objects and a central object of interest. This similarity is used to determine the position of each object within the cloud. However, this does not capture the semantic relationship of all the objects and the lack of consistency may reduce the expectation of finding an object when performing visual search. To generate a semantic 3D object cloud, we define and subsequently minimize an energy function that captures the pairwise similarity amongst all objects within the cloud. The energy is minimized using several statistical machine learning techniques and we show that the generated layouts from such techniques outperform those of other algorithms on a variety of metrics for evaluating layouts.